Status
Measured from every request the API served, including our own probes that run every few minutes. Nothing here is set by hand.
Degraded performanceUpdated Thu, 29 Oct 2026 09:13:08 UTC
Web search/search
- Success
- 81.88%
- Latency
- p50 369ms / p95 906ms
90 days agotoday
Images/images
- Success
- 88.06%
- Latency
- p50 537ms / p95 1109ms
90 days agotoday
Videos/videos
- Success
- 87.29%
- Latency
- p50 443ms / p95 1012ms
90 days agotoday
News/news
- Success
- 88.40%
- Latency
- p50 438ms / p95 1222ms
90 days agotoday
Shopping/shopping
- Success
- 88.54%
- Latency
- p50 764ms / p95 1463ms
90 days agotoday
Places search/places
- Success
- 100.00%
- Latency
- p50 478ms / p95 890ms
90 days agotoday
Map/map
- Success
- 100.00%
- Latency
- p50 476ms / p95 965ms
90 days agotoday
Place/place
- Success
- 99.96%
- Latency
- p50 210ms / p95 471ms
90 days agotoday
Reviews/reviews
- Success
- 100.00%
- Latency
- p50 557ms / p95 2052ms
90 days agotoday
Photos/photos
- Success
- 100.00%
- Latency
- p50 555ms / p95 2050ms
90 days agotoday
Autocomplete/autocomplete
- Success
- 100.00%
- Latency
- p50 53ms / p95 89ms
90 days agotoday
Patents/patents
- Success
- 100.00%
- Latency
- p50 1121ms / p95 1749ms
90 days agotoday
An endpoint is degraded when more than 5% of requests in the last 15 minutes failed with a server error, and at least 2 requests failed. It is down past 50% server errors. These are the same thresholds that page us.
Success is the share of requests that did not fail with a server error. Success and latency are measured at our edge over the last 24 hours.